Glioblastoma (GBM) is the most common primary brain tumor in adults. Standard of care includes neurosurgical resection, followed by radiation therapy (RT) and temozolomide (TMZ) chemotherapy. Although RT is one the most effective therapies for this aggressive disease, attempts to improve survival using RT dose escalation prior to the TMZ era have been largely unsuccessful. In one study, GBM patients treated with either standard dose radiation (60 Gy) or a range of escalated doses (up to the equivalent of 75 Gy), along with TMZ, showed no significant differences in recurrence or survival; however, this might be explained by dose escalation being limited to contrast-enhancing portions of the tumors in standard magnetic resonance imaging (MRI), which may not accurately predict tumor at high risk for recurrence. We hypothesize that radiation dose escalation may be beneficial, if all regions at high risk tumor can be identified and targeted. We propose to supplement standard MRI with proton spectroscopic magnetic resonance imaging (sMRI), a special technique performed to measure endogenous metabolite levels within defined volumes of tissue without having to inject a contrast agent. We conducted a clinical trial (R21 CA186169) by co-registering whole-brain sMRI metabolite maps with surgical planning MRI and imported into a neuronavigation system to guide tissue sampling in GBM patients. Samples were collected from regions with metabolic abnormalities in a biopsy-like fashion before bulk resection. Tissue samples were immunostained for glioma marker (SOX2) and analyzed to quantify the density of staining cells using an automated digital pathology image analysis tool. Correlations between sMRI markers and SOX2 density were evaluated. sMRI biomarkers exhibit significant correlations with SOX2-positive cell density. The choline to N-acetylaspartate (Cho/NAA) ratio showed significant associations with each quantitative marker (? = 0.82, p < 0.001). Intriguingly, sMRI metabolic abnormalities predated contrast-enhancement at sites of tumor recurrence and exhibited an inverse relationship with progression-free survival, which was published in recent Neuro- Oncology (Cordova et al, 2016). As it identifies tumor infiltration and regions at high-risk for recurrence, sMRI could complement conventional MRI to improve local control in GBM patients. Several other reports also demonstrated that a high Cho/NAA ratio correlated with areas of high tumor cell density and, after treatment, is specific for tumor recurrence. As such, sMRI could assist in defining the optimal volume to treat with higher, tumoricidal radiation doses. Our state-of-the-art sMRI is performed rapidly at high-resolution with 3D whole brain coverage. Our imaging would be of great value to radiation oncologists in choosing the optimal regions to target for dose escalation.